Radiotherapy is an image-guided process whose success depends critically on the imaging modality used for treatment planning and the level of integration of the available information. Currently, radiation treatment planning, done under the guidance of magnetic resonance imaging (MRI) or computed tomography (CT), is aimed at achieving geometrically conformal dose distributions within tumor volumes. While tumor biology plays a crucial role in the treatment outcome, neither conventional MRI nor CT provide such metabolic or biologic information, and efforts to deliver optimal treatments to cancer patients are thus compromised. Recent advancements in intensity modulated radiation therapy (IMRT) now make possible the precise delivery of deliberately non-uniform doses to better meet the differential requirements of spatially heterogeneous tumors as well as sensitive structures. Magnetic resonance spectroscopic imaging (MRSI) has also emerged as a powerful noninvasive tool for providing the type of metabolic information needed to identify biologically conformal dose distributions for improved radiotherapy. Working under the assumption that biologically-guided IMRT may ultimately achieve improved treatment outcomes, the Specific Aims of this technical development project are: (1) to develop a volumetric MRSI acquisition and data processing protocol optimized for use with radiation therapy treatment planning, (2) to establish a dose optimization framework which incorporates both metabolic MRSI and anatomic MRI/CT information, and (3) to demonstrate and evaluate the feasibility of achieving metabolically-conformal radiation doses through a small pilot study of 36 glioma patients. Successful completion of these aims will provide the tools required for subsequent clinical trials to establish the efficacy of MRSI-guided IMRT. While the results of this work are to be demonstrated on patients with brain lesions, the acquisition and processing methods also will be suitable to many other tumors including breast, head and neck, and prostate. In addition, the dose optimization framework will be equally applicable to biologically-guided IMRT incorporating functional information from other modalities including PET and molecular imaging.